Short-Term Changes in Skeletal Muscle Mass After Anthracycline Administration in Adolescent and Young Adult Sarcoma Patients

Abstract

Identification of anthracycline-induced muscle loss is critical for maintaining health in adolescent and young adult (AYA) cancer patients. We used routine chest computed tomography (CT) scans to investigate changes in skeletal muscle of 16 AYA sarcoma patients at thoracic vertebrae 4 (T4) after anthracycline treatment. CT images were examined at three time points (prechemotherapy, postchemotherapy, and 1 year). Significant changes in total skeletal muscle index and density were seen after chemotherapy (p = 0.021 and p = 0.016, respectively) and at 1 year versus baseline (both p < 0.05). This study supports the use of T4 as an early indicator of skeletal muscle loss in AYAs.

Sarcopenia, defined as loss of skeletal muscle strength and mass occurs with the aging process. In cancer patients, this process is accelerated and can occur at earlier ages.¹ Sarcopenia leads to reductions in physical functioning and the development of chronic health conditions (e.g., cardiovascular disease and diabetes) decades earlier than expected.^1,2 Unfortunately, sarcopenia is a common complication after treatment for young adult childhood cancer survivors^1,3 and has been considered a mortality predictor in various settings such as communities, nursing homes, and hospitals.⁴

Age, sex, lifestyle, and cancer therapies are risk factors for sarcopenia.⁵ Anthracycline-based chemotherapy, in particular, is implicated in loss of skeletal muscle mass resulting in sarcopenia.⁶ This is of special concern for the adolescent and young adult (AYA) cancer population (age 15–39 years) who receive anthracyclines as standard of care for multiple cancer types. To date, documentation of skeletal muscle mass and composition changes in AYA cancer patients remains sparse, with no current understanding of the short-term skeletal muscle loss and recovery resulting from anthracycline-based chemotherapy. Early identification of anthracycline-induced muscle loss and recovery is of vital importance to maintain the health of AYA cancer survivors who are at heightened risk for frailty and other chronic health conditions that may contribute to premature mortality. Given that sarcoma patients receive high doses of anthracyclines and chest computed tomography (CT) imaging is routinely used for disease response and surveillance, the purpose of this study was to investigate short-term changes in skeletal muscle CT imaging of AYA sarcoma patients at the level of thoracic vertebrae 4 (T4) after treatment with anthracyclines.

We identified 16 AYA patients (8 males and 8 females) who were treated within our center for sarcomas according to standard protocols consisting of combination chemotherapy with doxorubicin (cumulative mean dose 380 ± 77 mg/m²) and had available CT scans for analysis. This study was approved by the institutional review board. Leveraging sequential CT images, we measured indices of skeletal muscle cross-sectional area (skeletal muscle index, SMI) and composition (skeletal muscle density, SMD) for trapezius and rhomboid major (TRM), erector spinae and transversospinales (EST), and pectoralis major and minor (PMM) at T4 given its use in defining sarcopenia in other populations.^7,8 CT images before doxorubicin-based chemotherapy (baseline), immediately after the completion of the prescribed anthracycline (DOX) treatments (post-DOX chemotherapy), and 1 year since baseline were included in the analysis. After CT image selection at the anatomical level of interest, a web toolkit for analytic morphomics, CoreSlicer, was used to manually segment skeletal muscle areas for analysis.⁹ SMD was represented by Hounsfield units (HU). Fat infiltration or poorer muscle composition was expressed through lower HU values.¹⁰ The Hounsfield scale is conventionally centered on the density of water, 0 HU.¹⁰ Total SMI was calculated by summing $area∕heigh{t}^2$ for the three segments. Total SMD was calculated by summing the HU for each of the three segments, which was first multiplied by the segment area and then divided by total area $\left(\left(H{U}_1\times a_1\right)+\left(H{U}_2\times a_2\right)+\left(H{U}_3\times a_3\right)\right)∕TotalArea$. All statistical analyses were performed using SPSS software (IBM Corp, Version 21.0. Armonk, NY).

The mean age of the patients upon the first CT scan, before chemotherapeutic treatment, was 26 ± 5 years and the mean body mass index was 28.3 ± 12.3 kg/m². The average number of days between scan 1 and scan 2 was 151 ± 66 days. Of the 16 individuals, 8 had scans at 1 year ±24 days (4/8 were still on treatment at the 1 year scan). None of the patients received radiation therapy to the area of thoracic vertebrate T4. SMI was significantly lower post-DOX chemotherapy for both TRM (8.91 ± 2.93 cm²/m² vs. 7.7 ± 2.64 cm²/m², p < 0.01) and total muscle (23.97 ± 8.07 cm²/m² vs. 21.53 ± 7.18 cm²/m², p = 0.021) compared with baseline (Fig. 1). At 1 year, TRM and total muscle SMI remained significantly lower compared with baseline (TRM: 6.75 ± 1.87 cm²/m², p = 0.017 and total muscle: 17.49 ± 4.80 cm²/m² p = 0.011, respectively). Similar findings were demonstrated for EST, and PMM muscle groups at 1 year compared with baseline. Analysis of muscle composition revealed reduced SMD post-DOX chemotherapy for both TRM (42.18 ± 14.42 HU vs. 51.94 ± 11.07 HU, p < 0.01) and total muscle (39.07 ± 13.14 HU vs. 45.00 ± 12.82 HU p = 0.016) compared with baseline. At 1 year, TRM and total muscle remained significantly lower compared with the baseline time point (TRM: 40.32 ± 12.36 HU, p = 0.014 and total muscle: 37.17 ± 13.14 HU, p < 0.05, respectively) (Fig. 2). No significant differences were detected between post-DOX chemotherapy and 1 year. Results were similar at 1 year for patients who completed treatment and those who were still on therapy (p > 0.05, data not shown).

FIG. 1.

Skeletal muscle area quantified at T4 in adolescents and young adults with cancer. *p < 0.05 compared with baseline, ^†p < 0.10 compared with baseline.

FIG. 2.

Muscle tissue quality at T4 in adolescents and young adults with cancer. TRM, EST, and PMM. *p < 0.05 compared with baseline. EST, erector spinae and transversospinales; PMM, pectoralis major and minor; TRM, trapezius and rhomboid major.

We found, leveraging existing surveillance CT scans, that there was a significant reduction in skeletal muscle area and composition at vertebral level T4 after DOX chemotherapy, which was sustained at 1 year. This study is among the first to demonstrate skeletal muscle loss in AYAs during anthracycline-based chemotherapy. These findings are of key importance given skeletal muscle loss leading to sarcopenia and frailty is a marker of early mortality in AYAs and may be reversible by early detection and intervention.²

Conventional literature addressing cancer treatment-related skeletal muscle loss has predominantly focused on musculoskeletal late effects in survivors of childhood cancer and older adults.^11,12 When compared with age- and sex-matched normative populations, skeletal muscle mass reductions and impairments in limb strength and mobility have been demonstrated up to 20 years after childhood cancer treatments.³ Unfortunately, there is a paucity of literature, which has documented the deleterious effects of cancer treatment on skeletal muscle during and immediately post-treatment. To our knowledge, there is only one other recently published study that demonstrated skeletal muscle changes at the level of thoracic lumbar 12 (T12) using CT in children being treated for solid tumors.¹³ Furthermore, no study has focused on the AYA cancer population. The majority of AYAs will be cured of their initial cancer (5 year overall survival = 85%); as such, these patients are at high risk for treatment-related morbidity, further highlighting the importance of maximizing their short- and long-term health by understanding who is at greatest risk for skeletal muscle loss and intervening early to reduce muscle atrophy.

One of the greatest barriers preventing a systematic understanding of the trajectory of skeletal muscle loss and subsequent sarcopenia in AYA cancer population is the lack of research infrastructure to obtain detailed longitudinal measures of skeletal muscle mass and composition on a population scale. Existing approaches to estimate these are expensive, time-consuming, and burdensome to patients and providers. Overcoming these barriers will be essential to the development of targeted interventions aimed at mitigating skeletal muscle loss. CT of the chest is routinely performed in AYA cancer patients for disease response and surveillance. However, traditional quantification of skeletal muscle loss is typically investigated using T10-L5,^13–15 which may be outside of the standard chest CT imaging range and is more optimally seen in abdominal CT scans. Thoracic vertebrae T4, easily identifiable within a chest CT scan, has been highly correlated with muscle volume and the presence of sarcopenia in individuals with lung disease.⁷ Moreover, low skeletal muscle mass at T4 has been previously known to be a risk predictor of all-cause mortality in pulmonary patients.⁸

Our findings in this study using CT surveillance at T4 allows for future investigations on larger scale to understand natural trajectory and recovery of skeletal muscle in the AYA population. Future mechanistic studies are needed to guide and develop targeted intervention strategies to improve skeletal muscle function in AYAs. Moreover, given that we saw varying degrees of absolute muscle area loss and short-term recovery between the skeletal muscle groups, future studies should consider anthracycline type associated muscle wasting and recovery monitoring with regard to muscle specificity and histological fiber typing. Although all the patients within this study received high doses of anthracyclines, it should be noted that cancer diagnosis and tumor location varied. Predominant diagnoses included osteosarcoma, synovial sarcoma, and myxoid liposarcoma with 31% of patients experiencing lower extremity sarcomas. Moreover, individualized combination chemotherapies were used to treat the malignancies with doxorubicin and ifosfamide among the most common treatment regimens (56%). Cancer-related skeletal muscle loss is known to be multifactorial. Influences such as additional chemotherapeutic agents, tumor location, surgery, hospitalization, physical inactivity, and/or dietary habits should be considered as confounding factors in this study, which may exaggerate anthracycline muscle loss and, therefore, must be teased out in larger scale investigations. Finally, clinical trials to inform evidence-based approaches to prevent sarcopenia and decrease early morbidity and mortality outcomes in AYAs will be of key importance.

Ethical Approval

The study was approved by the IRB at the University of Texas M.D. Anderson Cancer Center.

Author Disclosure Statement

No competing financial interest exist.

Funding Information

This study was supported in part by a donation from the Texas 4000 and the National Cancer Institute P30CA016672 institutional core grant.